UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A search for variable radio emission in the galactic plane Taylor, Andrew Russell 1978

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1978_A6_7 T39.pdf [ 3.28MB ]
Metadata
JSON: 831-1.0094492.json
JSON-LD: 831-1.0094492-ld.json
RDF/XML (Pretty): 831-1.0094492-rdf.xml
RDF/JSON: 831-1.0094492-rdf.json
Turtle: 831-1.0094492-turtle.txt
N-Triples: 831-1.0094492-rdf-ntriples.txt
Original Record: 831-1.0094492-source.json
Full Text
831-1.0094492-fulltext.txt
Citation
831-1.0094492.ris

Full Text

A SEARCH FOR VARIABLE RADIO EMISSION IN THE GALACTIC PLANE by ANDREW RUSSELL TAYLOR B.Sc, University of Western Ontario, 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE DEPARTMENT OF PHYSICS We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1978 ^Andrew Russell Taylor, 1978 In presenting this thesis in partial fulfilment of the requirements f o r an advanced degree at the University of British Columbia, I agree t h a t the Library shall make it freely available for r e f e r e n c e and study. I further agree that permission for extensive copying o f this thesis for scholarly purposes may be granted by the Head o f my Department o r by his representatives. It is understood that c o p y i n g o r p u b l i c a t i o n o f this thesis for financial gain shall not be allowed without my written permission. n • * PHYSICS Department of The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 October 10, 1978 Abstract A description and the results of two surveys to search for highly variable radio emission in the galactic plane are presented. The surveys were carried out by making daily beamswitched maps of the survey regions and comparing the results of each day to the average of a l l days for evidence of variability. The f i r s t survey, at 21 cm, maps an area of 86 sq. deg. of the local spiral arm, in the direction of the constellation Cygnus, and searches for variability to a sensitivity of ^  0.3 Jy. The second survey, at 6 cm, covers an area of 56 sq. deg. in the galactic longitude range ( A 1 * ) 40°- 140°. The study of variability in this survey is not yet complete but variations at a level of ^  10 mJy are expected to be measurable. It is concluded that large intensity variations (greater than ^ 1 Jy) on a time scale of days are extremely rare; only one possible variation at a level of 0.9 Jy was detected during the 52 days of the 21 cm survey. A preliminary examination of the 6 cm data reveals one highly variable radio source, designated GT 0236+610, that reached a maximum flux density of 285 mJy. The results of further observations of this source are presented. The 6 cm observations were averaged over the complete survey period and the data were searched for compact sources down to a level of 10 - 20 mJy. The positions and 6 cm flux density for the resulting 196 compact sources are listed. i i i Table of Contents Page Abstract i i Table of Contents i i i List of Tables •' iv List of Figures and Illustrations v Acknowledgements V 1 Introduction 1 I The 21 cm Survey 5 A. The Survey 5 B. Reductions 8 C. Results 15 II The 6 cm Survey 2,1 A. Instrumentation and Methods 21 B. Calibrations and Sensitivity Limits 26 C. Reduction 39 D. Survey Results and GT 02 36+610 44 Summary and Conclusions 5 3 Bibliography 5 5 iv List of Tables Table Page I Summary of Results from the Accumulation Matrix 17 II 6 cm Calibration Sources 32 III Summary of Pointing and Gain Stability Analysis 33 IV A List of Compact Sources in the 6 cm Survey Region 45 V List of Figures and Illustrations Figure Page 1 The 21 cm Survey Raster 7 2 Cross-correlated Average Map of the 21 cm Survey Region 12 3 Figure of Merit versus Peak Signal to Noise 13 4 Cross-correlated Difference Map with 10 Jansky Test Source 14 5 Rms Confusion Level in Each Scan 19 6 Accumulation Matrix of 21 cm Variability Events 20 7 6 cm Receiver Schematic 24 8 Sample Sky Pattern of Telescope Motion 25 9 A Histogram of Survey Scan Confusion Levels 34 10 Beam Orientation with Respect to the Telescope Track 35 11 Source Offset versus Logarithmic Response Ratio 36 12 Beam Separation as a Function of Declination 37 13 Relative Beam Gains as a Function of Declination 38 14 Figure of Merit versus Rms Signal to Noise 43 15 Observations of GT 0236+610 at 2.8 cm 52 Acknowledgements I would f i r s t and foremost like to thank my supervisor Dr. P. C. Gregory for his guidance and support in a l l aspects of this endeavour. With regard to the 21 cm survey, I must also thank Dr. E. Argyle, not only for his aid in carrying out the observations but also for his hospitality, and the many stimulating discussions that made my stay at DRAG a pleasant one. Thanks must be given to the staff of both DRAO and NRAO for effi c i e n t l y dealing with the heavy demands this program made on the telescopes. Daryl Pawluk deserves recognition for his invaluable aid in developing the software to cope with the large amount of data produced. I must f i n a l l y thank Diane for patiently devoting many hours to the task of typing this manuscript. vii Dedication to Diane 1. INTRODUCTION Prior to the late 1960's i t was commonly assumed that, with the exception of the supernova remnant Cassiopeia A and some narrow band OH emission, no significant variations occured in the radio luminosities of discrete radio sources. Real source v a r i a b i l i t y was f i r s t established by Dent (1965) whose observations at 3.75 cm of the extragalactic, quasi-stellar source 3C 273 showed a substantial increase in flux density over the 3 year period 1962-65. By 1968 the l i s t of extragalactic radio sources showing variable emission at centimeter wavelengths had increased to 25 (Kellermann and Pauliny-Toth, 1968); a l l but one of these being either a Seyfert galaxy or QSS. In the span of l i t t l e over a decade, since the i n i t i a l discovery by Dent to the present time, the fact that radio sources exhibit variable continuum radio emission has become a well known and catalogued fact. The time scale for variation ranges from less than a second, for pulsars, to many years, for some extragalactic sources. In recent years a new class of highly variable, galactic radio source has been slowly accruing new members. Almost a l l of these sources are associated with binary star systems, some also being strong X-ray emitters. The highly variable nature of these sources is generally exhib-ited in the form of non-thermal, flaring radio emission, with a typical flare lasting anywhere from a few hours to a few days. An extreme example of the behavior of the known members of this class occured during September of 1972 when the infrared binary and X-ray star Cyg X-3 exper-ienced a radio outburst that lasted for approximately 9 days and reached the unprecedented peak flux level of 22 Jansky (Gregory et a l . , 1972a). 2. This gigantic outburst added a new, unexplored dimension to the possible intensity scales of radio source variations. The total energy emitted at radio wavelengths during this event was approximately IO1*0 ergs; compar-: -able to the energy emitted each second by a normal galaxy (Kellermann, 1974) but from a region with radius of the order of 101-Lf cm, or about 10 Astronomical Units (Gregory et a l . , 1972b; Gregory and Seaquist, 1974). In some cases high degrees of polarization have been observed during radio star outbursts (Seaquist et a l . , 1974; Owen et a l . , 1976). This fact, combined with the non-thermal nature of the spectra, strongly suggests that for some, i f not a l l , of these sources the emission is due to synchrotron radiation from r e l a t i v i s t i c electrons. The large out-bursts are then explained by the sudden injection of large quantities of energetic electrons into a region containing a magnetic f i e l d . This scenario may exist in close binary systems where sporadic, large scale mass transfer events can occur. This consideration together with the high percentage of binaries among the known highly variable sources indi-cates a possible relationship between the phenomenon of highly variable radio emission and some types of binary systems. Jaschek and Gdmez (1970) concluded that along the main sequence the percentage of spec-troscopic binaries remains essentially constant at approximately 50%. In spite of the fact that only a very small fraction of known binaries exhibit highly variable radio emission, many more sources may exist among the M.011 stars present in the galactic plane. With the exception of pulsars, which were f i r s t discovered acciden-t a l l y (Hewish et a l . , 1968) and for which systematic search techniques have been established, variable radio sources are, in general, detected 3. via one of two methods: (1) The regular monitoring of strong, discrete sources; previously detected during single beam, sky mapping surveys (Kesteven et a l . , 1976, 1977; Bridle et a l . , 1977). (2) Searches for radio emission from a specific class of astrophysical object (Owen et a l . , 1976; Geoffrey and Wallerstein, 1976; Spangler et a l . , 1977). At present, sky mapping surveys at radio wavelengths have been limited to source intensities greater than approximately 200 mJy and, since a given region of the sky is observed only once, are unlikely to detect a source which may be strong only a small fraction of the time. Thus, the f i r s t method entails the selection effect that generally only sources that are strong and relatively flux stable will be studied. The second method is biased by some assumption about what kind of object should be a radio source. These considerations motivated a decision to init i a t e a new type of survey aimed at answering the question: How time variable is the radio sky? Since variations on a conveniently short time scale are constrained, by light travel' time, to compact emission regions a beamswitching technique which has a well defined characteristic response to an unresolved source, and which is less sensitive than single beam radiometers to large scale extended structure i s ideally suited to this type of investigation. A survey to search for variable radio emission would consist of making re-peated beamswitching maps of a particular region of the sky. This dissertation describes two such surveys to detect variable radio emission in the galactic plane. The f i r s t survey, at 21 cm, covers an area 4. of about 86 square degrees centered on the Cygnus region of the plane. The second survey, at 6 cm, which covers a somewhat lesser area of 56 sq. degrees is the f i r s t stage of a complete survey of the galactic plane north of declination zero. A valuable, by-product of the 6 cm project, based on averaging a l l of the maps obtained for the var i a b i l i t y study, will be the f i r s t survey of the galactic plane for compact sources down to a level of 10+20 mJy; an improvement of close to a factor of 20 over that presently available. The major portion of this thesis will be devoted to the 6 cm survey for a number of reasons. First, the 21 cm survey is much less sensitive due to the smaller antenna size, higher system temperature and greater problem with confusion because of the larger beam size and separation. Second, previous results indicate that va r i a b i l i t y is generally larger at shorter centimeter wavelengths. Third, in many ways the techniques in-volved in the 6 cm survey were based on and evolved from the experience gained in the f i r s t attempt at 21 cm. Thus, the 6 cm survey is a second generation effort. The last reason, and possibly the most important, is that the 6 cm survey i s , as mentioned, the f i r s t stage of a program that wi l l cover the entire galactic plane above 6 = 0 ° . Hence this survey is part of an on-going project with a much wider scope both in terms of sky coverage, or completeness, and in the a b i l i t y to study var i a b i l i t y with time scales of both days and years. 5. Chapter I The 21 cm Survey A. The Survey This survey was conducted with the 25.6 meter, equatorial telescope of the Dominion Radio Astrophysical Observatory at Penticton, British Columbia, at a wavelength of 21 cm. The receiver system has a noise temperature of approximately 90 Kelvin and was operated at a bandwidth of 40 MHz, centered at 1.420 GHz. The receiver was connected to two prime focus feeds mounted at an orientation to produce two beams at common declin-ation. Each beam has a half power beamwidth (HPBW) of 36 minutes of arc and the beams are separated, through the symmetry axis of the antenna, by about 1.3 degrees. The antenna sensitivity at 21 cm is approximately 0.08 K/Jy. The receiver input is square wave modulated between the two feeds at a rate of 400 Hz. The receiver analogue output is sampled and digitized every 15 seconds, after passing through a postdetection RC f i l t e r c i r c u i t with a time constant t of 50 seconds. rc The theoretical receiver noise is calculated based on equation (1) (Tiuri, 1966). T AT = K 5X5— (!) r m s S / A A +-The postdetection equivalent integration time At is set to twice t and, since a beamswitching technique yields twice the signal to noise of the conventional Dicke switched system, the constant Ks has been assigned a value of one. The resulting rms receiver noise is 15 mK, which at 0.08 K/Jy 6. corresponds to a receiver sensitivity of about 0.19 Jy. In practice the sensitivity of the survey is only partially limited by this receiver noise level. The high concentration of intense sources in the survey region, coupled with the instability of the telescope pointing due to the moderate winds which are not uncommon at the observatory site, leads to a level of confusion generated noise (see section C.) comparable to, and in some regions greater than, this limit. The survey took place over three approximately 10 day periods and one 21 day period during the summer of 1977. Each days observations consisted of mapping an area of 86 sq. deg. looking down the local spiral arm of the galaxy in the Cygnus region. The map was accomplished by means of a raster arrangment of 39 scans at constant declination, driven eastward at the sidereal rate (see Figure 1.). Each scan is 22.5 minutes in dur-ation, spanning 45 minutes of right ascension. The scans are separated by 15 minutes of arc, or one-half of a HPBW, in declination. The starting right ascension i s augmented by one minute after each scan so that the raster remains centered on the galactic plane. Observations begin each day at the same hour angle (and hence sidereal time) to reduce variable spill-over effects. The receiver gain is calibrated during the data re-duction process (see section B.) but as a further check the gain is mon-itored by f i r i n g a stable noise tube at the front end of the receiver system prior to each days observations. The telescope operation and data aquisition is controlled by a PDP 11/03 minicomputer. The telescope control software, originally designed by Dr. J. Gait, was slightly modified in adaptation to this project. The receiver output was read every 30 seconds of RA and temporarily stored on floppy disc for later transferal to a PDP 11/45 for final processing. u: C 42 40 36 h -34 r 32 h~ BEAM SIZE GALACTIC PLANE V \ •v 3 £ \ | T l X iCTQ - \ • C 3 8 f - \ |° • - \ ~ -3 ^ 3 ^ SCAN 1 21 h 50 m 40 m 30 m 20 m 10 m 20 h 50™ 40 m RA 8. B. Reduction The software for processing the 21 cm data was i n i t i a l l y developed by Alex Szabo, under the direction of Drs. P. C. Gregory and E. Argyle, during May of 1976 in connection with an earlier attempt to survey the Cygnus region. Although the original flow of this program remains essent-i a l l y intact, extensive changes have been made, by Dr. E. Argyle and myself, in u t i l i z i n g the program in this experiment. The primary function of the program is to search for and measure differences, with the characteristic signature of an unresolved source, of a single days observations from an average created with data taken over a number of days. Some caution needed to be taken in computing an average from a given set of days to ensure that data which is contaminated by interference or unacceptable due to faulty operation of the telescope or receiver system is not included. A preliminary average is f i r s t constructed for a specific scan using data from a l l the days to be processed. The individual, daily observations are then calibrated to match the gain of the average scan. This is done by calculating a multiplicative constant which, when applied to the data of a given day, minimizes the rms energy in the difference from the average. This scheme of normalizing the observations internally, r e l -ative to the mean gain over a number of days, takes advantage of the presence of strong, stable confusing structure in the data and is specifically suited for va r i a b i l i t y studies since the total system, including the ant-enna efficiency and the receiver gain, is calibrated in a single process. The resultant gain factor corresponds to an integrated value covering an approximately 25 minute period. After calibration, the rms energy in the difference array for each day is compared to the mean value over a l l days. 9. When the r a t i o exceeds an assigned l i m i t (usually taken to be 1.4) that day i s rejected from the average. A new average i s then computed and the pro-cess i s iterated u n t i l no rejections occur., The adopted rejection r a t i o of 1.4 resulted i n an average rejection of about 25% of the data. A contour map of the survey region constructed from an average of 11 days observations, cross-correlated with an instrumental response to an unresolved source, i s shown i n Figure 2. The contour values range from zero to a maximum of about 135 Jy. Once a satisfactory average has been established and the d a i l y observations normalized to the mean gain, the program searchs for v a r i a -tions from the average i n the difference arrays. The difference array i s f i r s t cross-correlated with an instrumental p r o f i l e and then scanned for positive and negative extrema. When a peak i s found the shape of the signal at the corresponding location i n the uncorrelated difference array i s used to calculated a figure of merit for the suspected variable. The figure of merit i s a measure of the s i m i l a r i t y between the shape of the signal p r o f i l e and the instrumental response. Its calculation-; was designed i n such a manner as to be 100% for a true unresolved source i n the absence of noise. I f the signal i n the region of the suspected variable i s desig-nated by a function S(t)'", and the instrumental p r o f i l e by P ( t ) , the figure of merit may be expressed as i n equation (2). F - 1 0 0 - [ 1 - < S W - I , P ( ^ > ] (2) < S(t) > Diagonal brackets are used to indicate the root mean square value of the argument over the range of an instrumental response. The constant I scales the instrumental response to the signal intensity. It i s obvious from inspection of equation (2) that the figure of merit i s an estimate 10. of the fraction of the signal energy residing in a source-like profile. The signal is accepted as a possible variable source i f the figure of merit exceeds 50%. The change in the figure of merit as a function of signal to noise was investigated by inserting a r t i f i c i a l l y constructed sources at various intensities into the real data prior to the search portion of the program. The results are plotted in Figure 3. The imposed limit of 50% on the figure of merit effectively limits output, in the case of real source variat-ions, to a signal to noise of about 1.5 or greater. A strong, source variat-ion could be identified on the basis of a single days results both because of the high figure of merit and because, due to the proximity of the scans in the raster configuration, confirming signals would be produced at the same location in two or more adjacent scans. An example of the response of the reduction system to a strong variable is shown in Figure 4. which is a map of the cross-correlated difference of one day from the average with a 10 Jansky source injected into the observations. Small scale variations, which appear in one scan only, must occur on more than one occasion to be considered of celestial origin. An over-all perspective of the positions and frequency of occurence of these weaker variations is provided by a simple matrix display of the search output. Each element of the constructed 39x90 matrix corresponds to a unique scan number and data sample. Upon detection of a possible variable by the search rout-ine the matrix element defined by the position of the detected peak is augmented by one. Weak variable sources which may vary on a time scale of a few days or less would appear as an accumulation of events at a specific matrix element. The significance of any such accumulation is estimated by computing the probability of the chance ocurrence of an accumulation of equal magnitude, assuming random variations in the data due to system ins t a b i l i t i e s . The probability of exactly n random variations ocurring at the same matrix element is given by the Poisson approximation to the Binomial Distribution (Feller, 1950). . n -A POO = (3) Here X is the mean occupation number of a matrix element given by N/T, for N total v a r i a b i l i t y events in a matrix of T elements. The frequency of ocurrence f(n) of accumulations of magnitude n, assuming random place-ment into the matrix is then given by T*P(n). Some non-random process is assumed i f the observed frequency is greater than approximately three times the Poisson frequency. Figure 2. Cross-correlated Average Map of the 21 cm Survey Region , - H O CTlOOI^vO LO " v J - t O Tj- t*0 to rO fO to to to (saayoaa) NOiivNnoaa Figure 3. Figure of Merit versus Peak Signal to Noise 10 15 PEAK SIGNAL TO NOISE RATIO 14. 15. C. Results A report of the results of this survey would be incomplete without some discussion of the survey sensitivity limits. As has been mentioned (see section A.) the rms noise of the receiver system of 15 mK corresponds to a system sensitivity of about 0.19 Jy. In the absence of confusion generated noise the signal to noise factor of 1.5 introduced by the figure of merit c r i t e r i a sets the effective lower limit on measurable variations at about 0.3 Jy. Subject to some limit on the stability of the telescope and receiver systems, the process of differencing two scans of the same region of the sky on separate days w i l l leave a residual energy in the result; the magnitude of which is dictated by the level of confusion in the observations. The rms confusion level in each scan of a 21 day average of the survey region is plotted in Figure 5. With gain matching, differen-cing was found to yield an rms noise level, in excess of receiver noise, equal to about 0.6% of the rms confusion level. In the worst case (see Figure 5. scans 33 and 34) the detection limit is as large as 1 Jansky, but for about 80% of the survey the minimum detectable variation i s close to the lower limit of 0.3 Jy. The survey time coverage includes the following dates in 1977: May 6-> May 20; June 20->June 29; July 8-hJuly 21; August 31->September 20. Taking into account days lost due to equipment breakdowns the total time coverage of the survey is 52 days. During this time only one signal was produced that satis-fied a l l the detection c r i t e r i a established in the previous section. The sig-nal occured on September 13, 1977 in the f i r s t scan of the raster at a level of approximately 0.93 Jy and with a figure of merit of 82.1%. The detection was confirmed by the presence of a 0.32 Jy peak in the adjacent scan at the 16. same right ascension. Based on these two results and knowledge of the beam shape a variable source producing these signals would be situated approxi-mately 5 minutes of arc south of the f i r s t scan of the raster at: a(1950) = 20 h 17 m 54 s ± 15S 6(1950) = 31° 50' ± 5' The rms confusion level in scan 1 has the relatively low value of 3.02 Jy. The a b i l i t y to routinely subtract out confusion to a level of 0.6% coupled with a lack of structure in the average at the location of the detec-ted signal, seems to rule out the possibility that the output was created by gain or pointing fluctuations. A signal was detected again at the same raster position two days later, on September 15, at the slightly lower strength of 0.83 Jy. On this occasion no output was produced at the corresponding location in the adjacent scan. However, the intensity of the signal in scan 1 taken together with the results of the earlier detection would place any response in scan 2 below the figure of merit threshold. Thus, this latter event, although not significant in i t s e l f , is consistent with the possible presence of a highly variable radio source at the derived co-ordinates. Further measurements, with greater sensitivity and resolution, are required to examine this possibility. Figure 6. shows the matrix output for the 21 day observing period August 31-*September 20. A total of 590 variations in the data were recorded. The observed frequency of occurrence of accumulations of a specific size and the corresponding Poisson frequency ( see the previous section) are listed in Table I. Accumulations of up to 2 events per matrix element do not de-part significantly from that due to random placement into the matrix, as expected for variations caused by small system i n s t a b i l i t i e s . However, the Table I: Summary of Results from the Accumulation Matrix (1) (2) (3) Magnitude of Observed Poisson Accumulation Frequency Frequency 0 2997 2972 1 461 494 2 35 41 3 14 2 4 0 .1 5 2 .003 6 0 10"6 7 1 10"8 18. large deviations from the Poisson distribution for accumulations greater than two clearly indicate that some non-random factor is causing a high degree of clumping of the variations around certain regions in the sky. Further investigation showed that these large accumulations are associated with high gradients in the confusion structure and hence can be attributed to fluctuations in the telescope pointing, which were typically about 2-3 minutes of arc. The results of this survey indicate that radio outbursts of the c a l i -ber of those observed from Cyg X-3 are extremely rare. The absence of var-iations greater than 1 Jy in the 86 sq. deg. area and 52 day time coverage of the survey implies an upper limit on the density of radio flare events above 1 Jy, in this region of the galactic plane, of 5 X 10 sources ster •'•-day 1. One possible variation was observed with a magnitude of ^ 0.9 Jy. The position of Cyg X-3 is in a highly confused region of the survey, making an estimate of i t s flux density impossible. Only the crude upper limit of ^1.5 Jy may be placed on variations of Cyg X-3 during the period of this survey. Analysis of the matrix display does not rule out the possibility that variability below ^ 0.5 Jy may be more common. The sensitivity of the equipment allows l i t t l e light to be shed on this interesting possibility. The resolution of this question requires a similar survey with better sen-s i t i v i t y and resolving power. Figure 5. Rms Confusion Level in each Scan. 19. 3 • • • • —I 1 1 1 I i I 10 20 30 SCAN 20. Figure 6. Accumulation Matrix of 21 cm Variability Events SCAN 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 7 - 1 1 1 8 1 • • • • 1 • • • 1 1 2 5 2 1 • • 9 2 1 • 1 • • • 1 • • • • 2 • • • 10 1 • 1 • • • 1 1 1 1 • • 1 11 1 1 1 - 1 2 • • • • 12 1 1 . . . . 13 1 1 • • 1 • • 1 • • 1- • 1 14 1 1 1 • • 1 15 1 2 1 1 16 • • • 1 2 1 1 . . . . 1 17 • • • • 3 1 1 . . . l . . . 18 • • • 7 • • • • 1 • • • • 1 1 19 • 1 2 1 2 1 • • 1 1 • • • 1 • • 20 • • 1 2 22 • • 23 • 1 25 • 1 26 • • 28 2 2 29 • 1 • • • • 3 3 31 • • 3 1 • 1 3 1 1 32 1 1 2 33 2 1 • • • • 34 1 2 1 1 - 1 1 1 1 2 • • 35 36 • -. 1 1 • 1 • • 2 • • 1 2 1 • • 1 37 • • 1 3 i . . . 38 • • • • • 1 2 1 • • 1 • 1 1 . . . 39 40 • • • 1 1 • • • 4J 1 1 - • - - 2 • 42 1 - 2 43 • • • 1 • 2 1 1 44 • • • 1 • • • . 45 1 • , 2 1 • • • 1 46 . . . l . . . . 4 7 - 1 1 2 49 50 51 52 1 1 1 1 • • • 1 • • • 1 - 1 1 -1 1 54 1 • • 55 1 1 • 56 1 • 1 57 • • • 58 • • • 59 • 1 1 61 -. 1 1 1 1 1 I l l -62 1 2 1 63 1 1 • • 1 64 1 1 • • 1 1 • • • 2 • • 1 1 • • • 65 1 1 . . . 1 3 1 1 66 1 1 1 - - 1 - - -67 1 • 1 . . . l l 1 - 2 - - - 1 - - -6 8 - 1 1 - 1 1 69 1 • • 2 1 • • 2 70 1 1 • . . . l 1 1 71 1 • • 1 • • • • 1 1 73 • 2 • 1 • 1 1 1 1 1 • • 2 1 • 1 • 1 74 • • 1 1 • • 1 1 - 1 • 2 1 1 • • • 75 • • 1 • 1 • • • 1 2 3 76 • • 2 1 • • • 1 1 1 • • • 1 • • • 77 1 • . . . 1 1 3 • • • 1 1 1 1 • 1 • • • 78 - 1 - • - 2 - • - 1 1 - 2 - - 1 2 1 1 - 2 - 1 2 • • • 79 2 - 2 : . . . i i . . . i 80 2 1 • 1 1 . . . 1 . 1 1 1 . 1 1 . 1 1 . . 1 . . . 1 . 1 1 . . 8 1 - 1 1 1 . . . 1 82 1 1 1 1 - 1 1 1 - 3 1 -83 1 • • 1 • • 1 2 1 • • • • 1 • 1 2 • • 84 • • • • 1 1 1 . . . 1 . . . i 3 85 1 86 2 87 2 88 89 1 . . . l 90 21. Chapter II The 6 cm Survey A. Instrumentation and Methods A l l the observations were carried out with the 91 meter transit telescope of the National Radio Astronomy Observatory at Green Bank, West Virginia, at a wavelength of 6 centimeters. The telescope receiver system is illustrated schematically in figure 7. Each of the two cooled parametric amplifier receivers has a total system temperature of 66 K and a 3 db bandwidth of 580 MHz. The receivers are connected to two feeds located on a rotatable mount at the prime focus of the telescope. The feeds are offset in such a way as to produce two beams on the sky, separated by 7.2 minutes of arc through the telescope axis of symmetry, each with half power beamwidth of 2.7 minutes of arc. The beamwidth and separation were found to be weak functions of declination; both quantities having a minimum at the zenith. The telescope sensitivity at the zenith is approximately 0.9 K/Jy. The radiometer system was operated in the beamswitching mode, in such a manner that receiver 1 is switched between feeds A and B 180 degrees out of phase with receiver 2, at a rate of 50 Hz. The dual receiver system thus yields two independent measurements of the differential output of feeds A and B; greatly enhancing the r e l i a b i l i t y of transient source detection and providing a /2 improvement in receiver noise when combined. In addition to removing the effects of variable atmospheric emission, beamswitching with two closely spaced beams greatly reduces the problem of noise generated by the confusion generally associated with '22. observations in this frequency region at low galactic lattitudes. Con-fusing sources with angular extent greater than approximately 8 minutes of arc are effectively filtered out by this technique. The survey region, which consists of a portion of the galactic plane in the galactic longitude range (&^) 40 to 140 degrees and extending approximately ± 2 degrees from the plane, was mapped once each day during the twenty day period, August 11 -> 30 of 1977. Since the transit tele-scope i s , in effect, constrained to motions in one dimension considerable undersampling occured in RA during the mapping operation. Hence, although the survey encompasses a region in the sky of approximately 400 square degrees only 56 square degrees or 14 % of the total area was actually covered by the observations. Figure 8 shows the sky pattern of the tele-scope motions for a sample portion of the survey. The telescope is driven along the meridian at a rate of 130 minutes of arc per minute of time. Each scan is two minutes in duration or 4.3 degrees long and is centered on the galactic plane ( b 0 ° ) . Due to earth rotation the scans are inclined to a great cir c l e of constant right ascension by an angle $ which varies with declination in the approximate range 3.0 to 6.5 degrees. Con-secutive scans are driven in opposite directions and are separated by approx-m s imately 2 20 in right ascension. After each northbound scan the telescope is held stationary for 12 seconds and during this interval a stable noise tube is fired at the front end of the receiver system. In this way the gains of both receivers are calibrated once every five minutes during the entire mapping operation. At a driving rate of 130'/min a HPBW of 2.7 minutes of arc corresponds to a temporal resonse, to an unresolved source, of characteristic frequency 23. f equal to 0.8 Hz. In order to be able to distinguish high amplitude impulse noise from the slower source response the post-detection low pass f i l t e r has a 3 db cutoff frequency f extended to 3f Q. To prevent aliasing from the higher harmonics of the sampling frequency into this bandpass the radiometer analogue output is sampled at a frequency of 2 f c > This corre-sponds to a sample time of 0.2 seconds, during which time the telescope wi l l travel 26 seconds of arc or 1/6 of a beamwidth along the sky track. The integration period, given by 1/3 f , is 0.4 seconds. With a 580 MHz bandwidth the theoretical rms noise fluctuations in the radiometer output given by equation (1) with Kg= 1 and At = 0.4, is 4.3 mK. This noise level is considerably reduced by additional post-detection di g i t a l f i l t e r i n g and by combining the output of both receiver systems. The data from both receivers are written by an on-line computer onto a magnetic tape which is later transported to the NRAO computer f a c i l i t y for i n i t i a l processing. Each days observations results in 188 survey scans and 94 of the 12 second receiver calibration-scans from each receiver. A survey scan contains, i n addition to the six hundred data samples, informa-tion concerning the telescope pointing, the receiver gains and the scan and receiver identifiers. The analogue output of both receivers is monitored in real time at the telescope via strip chart recorder. Output which is unusable due to either interference, faulty telescope performance or poor weather conditions is noted at this stage. TO SWITCH DRIVE OSCILLATOR * > > Noise Cal "* 1 Noise Cal COLD LOAD TO FRONT END SWITCHES 1 2 3 • • » "A" PARAMP 1 2 3 a » • "B" PARAMP TRANSISTOR AMPLIFIER 4.5 to 5.1 GHz FILTER LEVEL PUMP COLD LOAD LEVEL PUMP L. 0. AMP X2 < L. 0. TRANSISTOR AMPLIFIER 4.5 to 5.1 GHz FILTER IF AMP SQ. LAW DETECTOR SYNC. DETECTOR SWITCH DRIVE OSCILLATOR MIXER IF AMP ~ T ~ SQ. LAW DETECTOR SYNC. DETECTOR LOW PASS FILTER "A" out — f e -MIXER LOW PASS FILTER "B" ou 4^ 25. Figure. 8 Sample Sky Pattern of Telescope Motion J I I I L OO \ 0 L O ^3" (soa-iSoa) (0S6I) N0IlVNIlD3u° 26. B. C a l i b r a t i o n s and S e n s i t i v i t y Limits 1. Considerations. The question of the s e n s i t i v i t y of the survey i s two-fold since the data may be interrogated f o r two types of behaviour. The primary aim of the survey i s the detection of highly v a r i a b l e sources which are gener-a l l y absent, i n that any quiescent emission i s below detectable l e v e l s , but sp o r a d i c a l l y f l a r e up for periods of a day or more. The s e n s i t i v i t y with regard to t h i s type of phenomenon i s given by the minimum detectable f l u x l e v e l on a single days observation. This quantity i s determined e m p i r i c a l l y from the rms receiver noise and any noise generated by confusion. Using a technique, s i m i l a r to that of the 21 cm survey, of taking differences from an average scan the confusion generated noise w i l l be some small f r a c t i o n of the rms confusion l e v e l i n a si n g l e days observations. The l e v e l of confusion varies from place to place along the g a l a c t i c plane. In the Cygnus region, where a large number of intense, compact sources e x i s t , the rms confusion amplitude has i t s largest value of approximately 200 mK. This compares with ^ 80 Jy f o r the 21 cm survey, i n d i c a t i n g the value of c l o s e l y spaced, narrow beams. Fortunately, for the major portion of the survey, p a r t i c u l a r l y i n the region of the g a l a c t i c anti-center, the confusion l e v e l remains f a i r l y close to the mean value of 7 mK. Figure 9. shows a h i s t o -gram of observed confusion l e v e l s i n the survey scans, excluding the contribu-t i o n from scans through the Cygnus region. As mentioned, the rms receiver noise i n a s i n g l e , untreated scan i s t y p i c a l l y about 4 mK. During the data reduction process the scans are cross-correlated with a gaussian approximation to the beam shape. This procedure reduces the noise l e v e l to about 1 mK. Thus even with the conservative estimate of 10% confusion i n the d i f f e r e n c e , f o r the major portion of the '27. survey a three sigma detection limit of 5 mK is attainable. The corresponding flux density range is 6 mJy at the observatory zenith to about 18 mJy at zero declination. The more detailed measurement of the daily variations of these sporadic sources, when they occur, and of those sources observed to be present on a l l days of the survey involves an additional error which is proportional to the source intensity. This error manifests i t s e l f as residual source variations after the data has been processed. The mean amplitude of these residual variations reflects the degree of sta b i l i t y of the telescope over a 24 hour period and ultimately sets the effective sensitivity limit for measuring variability of strong sources. For this reason one of the chief concerns in ini t i a t i n g this survey was to define, and when possible minimize, these r e s i -duals. The problem is assumed to be due to two separable component variations in: (1) the telescope pointing, (2) the telescope antenna gain. Small fluctuations in the telescope pointing from day to day can lead to large changes in the radiometer output, particularly i f the source traverses a portion of the beam pattern with a large gradient. In order to be able to distinguish this type of behaviour from the source variations, knowledge of the source position in the direction orthogonal to the sky track of the tele-scope i s required. This is achieved by rotating the telescope beams at a fixed angle to the track, as shown in Figure 10. The component of the beam separation normal to the track is equal to one-half of a HPBW. The radiometer response as the telescope scans through an unresolved source is approximately sinusoidal in shape with, in general, unequal positive and negative excursions. The ratio of the peak positive response to the peak negative response Rg is independent of source strength and is a unique function of the perpendicular 28. offset 6 of the source from the telescope track. This method of observing as opposed to aligned beams, in addition to allowing more accurate determin-ation of source position and flux density, increases the effective sky cov-erage of the survey by 50%. In order to exactly determine the response ratio as a function of the source offset the relative gain of the two beams should be mapped over the entire response region and at a number of declinations. A complete beam mapping operation on this scale was considered to be too time consuming to be feasible, therefore a reasonable compromise was adopted. Drift scans through standard calibration sources, with the beams rotated to a position angle of 90° from the meridian, were used to obtain accurate cross-sections of the beam shapes over a wide range of declinations. The response ratio functions were then computed assuming eylindrically symmetrical beams and using only the outer portion of the profile where there is minimal distortion of the beam shape due to the presence of the other beam. The resultant calcu-lated source offset is plotted against the logarithm of the response ratio for three declinations in Figure 11. For small offsets (less than approximately 1') the data may be represented to a good approximation by a f i r s t order polynomial, as predicted by a gaussian model of the beam profiles.- The varia-tion in the slope reflects the declination dependence of the beam shapes and separation (see Figures 12. and 13.). The value of the gains of the two beams at the derived source offset position is used to normalize the positive and negative response to a gain of one.. Source intensities determined in this manner wi l l be independent of daily fluctuations in the telescope pointing. Variations in the antenna efficiency are indistinguishable in the survey data from true source variations and appear 29. to occur on a time scale of hours or less, making the effects d i f f i c u l t to calibrate out using sources of known flux st a b i l i t y . Thus with the successful removal of variations due to pointing changes the magnitude of the antenna gain fluctuations define the limit to which residual variations may be reduced. In an effort to minimize this limit the pre-caution was taken of carrying out the observations at night to avoid the effects of variable solar heating on the physical superstructure of the telescope. 2. Measurements. The st a b i l i t y of the antenna gain and the consistency of the telescope pointing were monitored by scanning through the calibration sources 3C 48 \ NGC 7027 and DR 21 on each day in a manner identical to the survey observations. These calibrators were selected primarily for their stability, small angular size, relatively high intensity at 5 GHz and prox-imity to the galactic plane. The properties of these three calibrationr sources are listed in Table II. Daily responses from these sources are averaged over both receivers and then over a l l days and the resulting response ratio R./RD is used to calculate A D a mean offset position 6 for the complete observation period. A small differ-ence in the telescope pointing from the mean will produce changes of a di f f e r -ent sense in the two beam responses. The magnitude and sign of the variation is a function of the mean offset i t s e l f , since this defines the slope of the gain curve, and the size and direction of the pointing change. The fractional variations in the responses due to changes in the antenna gain wi l l be common to both beams. This fundamental difference in the behaviour of the component variations allows their separate contributions to be extracted from the total 1. Kesteven et a l . , 1976 classify 3C 48 as a possible weak variable at 2.7 GHz. 30. residual variations in the calibration sources with a relatively simple analysis of the data. The maximum response of the radiometer to an unresolved source of strength S as i t passes through one of the beams will be given by the relation: R = S-G(0) (4) Here G(0) is the gain curve of the beam as a function of the perpendicular offset. Assuming that the source strength remains constant then the daily excursions of the peak response from the mean must reflect changes in value of the gain function only. This can occur through variations in the int r i n -sic amplitude of the function, corresponding to changes in the antenna gain, or variations in the argument 0. To f i r s t order i t is convenient to adopt a mean value of the gain function'G given by the measured beam profile and the mean offset 0 of the calibrator. The total variations in the gain func-tion including the effects of both antenna gain and pointing is independent of source strength and can be measured directly via the equation: AG = (G-G) = AR-G (5) R This gain variation can be measured for each beam independently and the results for any given day must satisfy the relation: AG = A0-(^|)_ + Ag-G (6) 0 dG Here ("jgO_ 1 S the slope of the gain curve and under the assumption that A6 0 is small the value at 0 is used. The quantity Ag is the fractional antenna gain change. The variations A0 and Ag are common to both beams thus the pointing variation can be extracted by isolating Ag in equation (6) and 31. differencing the results of beams A and B. A0 = B B / dG, 1 dG_ 1 "I A B d6 G A d9 G (7) The resulting value of A6 can then be used in equation (6) to solve for the antenna gain variations. This calculation is done independently with the output of both receivers to check the consistency of the solution. The results of this analysis for the three calibration sources used are summarized in Table III. Column 1 contains the rms fluctuations, in arc seconds, of the telescope pointing in the offset direction. These numbers represent variations from the mean offset only, and thus do not reflect the quality of the absolute pointing of the telescope. The rms antenna variations are listed in Column 2. No correlation exists between measurements made on the same day with a l l three sources, indicating that the variations occur on a time scale of a few hours or less. For sources that produce outputs with sufficient signal to noise in both beams that accurate offsets can be obtained from a single days observation, the two sigma value of approximately 4% on the antenna gain fluctuations represents a good estimate of the limit on reliable measurements of source variations. The rms of the total residual variations for each beam due to both pointing and gain effects are listed in Column 3 of the table. Table II: 6 cm Calibration Sources (1) (2) (3) (4) (5) (6) (7) Source cc(1950) 6(1950) S(6cm) Angular Size £ H b 1 1 Jy DR 21 20 h 37™ 14?2 42° 09' 0675 21. < 20" 48?7 0?5 NGC 7027 21 05 09.4 42 02 02.8 5.75 1"- 5" 51.9 -3.5 3C 48 01 ,;34 49.8 32 54 20.7 5.37 < 1" 101.0 -28.7 33. Table III: Summary of Pointing a Gain Stability Analysis 1 2 3 Source Pons Variation Rms Antenna Total Rms Source Variation (%) in Offset (") Gain Variation A . B DR 21 3.2 0.8 1.1 3.5 NGC 7027 2.4 1.8 1.7 3.9 3C 48 1.2 1.7 1.9 2.2 •'34. Figure 9. Histogram of Survey Scan Confusion Levels. o CN z o O l-H •<r* CO H, z - o CO 00 • vO o CN LO O SNVDS do yaawnk gure 10. Beam Orientation with Respect to the Telescope Track Figure 11. Source Offset versus Logarithmic Response Ratio X 6 = 34° 0 + 6 = 44° 6 = 70° O O x 9 -I I I I I I I -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 SOURCE OFFSET (minute of arc) Figure 12. Beam Separation as a Function of Declination 10 20 30 40 50 60 70 80 DECLINATION (DEGREES) 38. Figure 13. Relative Beam Gains as a Function of Declination 1.2 CO Z i—i < < o 1.0 20.9 < OS 0.8 • • I I I I I L_ I I I l _ 10 20 30 40 50 60 70 80 DECLINATION (DEGREES) 39. C. Reduction Ini t i a l processing is carried out using the standard NRAO continuum data reduction program, called C0NDARE3. Phase I of this program is exe-cuted with the telescope produced tape upon arrival at the NRAO computing f a c i l i t y after each days observation. At this stage the data and header information from each scan is written on a user magnetic tape. A scan is uniquely identified by a source name, receiver number and date. The source name is the 1977 right ascension of the beginning of the scan and is thus shared by identical scans (through the same region of the galactic plane) from each day. An average cal factor i s computed for each of the 12 second receiver calibration scans and applied to the observations to convert the data to antenna temperature. For each set of scans associated with a given source name an average scan is produced, using the results from a l l days, and is displayed on a calcomp plotter. These plots are used to establish, by visual inspection, the regions suitable for baseline removal. The remain-der of the data reduction process i s carried out with the University of British Columbia IBM 370. A master f i l e is created which consists of a l i s t of a l l source names used in the survey. Along with each source name is written the epoch 1950 right ascension and declination for the beginning of the scan, a 20 element array containing a set of flags indicating the quality of the data on each day and specifications for up to three regions to be used for calculation of a baseline. This master f i l e is read by the main program when processing is i n i t i a l i z e d . The f i r s t program to be run on the data is concerned with the survey for low level compact sources^in the .observations. The routine simply 4.0. conducts a search for unresolved source profiles in the observations. "The standard procedure in the case of a beamswitched scan through an unresolved source is to cross-correlate the data with a beamswitching model of the instrumental response. In our case, because of the orientation of the beams with respect to the sky track symmetrical beam profiles are not generally expected so this method was not adopted. The raw data i s , instead, cross-correlated with a gaussian model of a single beam profile in order to in-crease the signal to noise for an instrumental profile with arbitrary beam ratio. After baseline removal the scan is searched for positive and negative peaks by locating the zeroes in .the f i r s t difference array. Estimates of the positions and intensities of the peaks are refined by quadratic interpolation using the three data points in the peak region. This technique for finding extrema is highly sensitive, so that, even though most of the high frequency noise is eliminated by the cross-correlation, a significant number of noise peaks wi l l be included in the search results. Further processing, at this point, reduces to the problem of f i l t e r i n g out those peaks which are due either to noise, extended sources or the complicated structure of confusion. The f i r s t step in distinguishing responses due to compact sources in the array of positions and intensities of the positive and negative peaks is to search for the appropriate beam configuration. For a northbound scan, in which feed A precedes feed B along the sky track, the instrumental profile consists of a positive peak followed by a negative peak. This beam relation-ship is referred to as the positive sense. Southbound scans have negative sense in that the negative peak occurs f i r s t . The measured beam profiles allow the expected peak separation, as a function of declination, (see Figure 12.) to be determined to within an error of one data sample for strong sources. For each positive peak the entire set of negative peaks in the same scan are searched and a possible source is flagged i f the relative positions f a l l within tolerance of the c r i t e r i a set by the sense and declin-ation of the scan. Before a possible source is input to the next stage of the program the peak to peak amplitude of the signal is calculated and the output is rejected i f the amplitude is less than 10 mK. In this way t r i g -gering of a source output due to random coincidence of two noise peaks is eliminated. As a fin a l test to establish whether a signal is due to an unresolved source, the shape of the signal in the region of the proposed source is com-pared to a constructed instrumental profile with the same positive to nega-tive peak ratio. A figure of merit is calculated using an expression identi-cal to that used for the 21 cm survey (see equation (2).). A r t i f i c i a l l y constructed sources were inserted into the real data to determine the variation of the figure of merit with the signal to noise of an unresolved source re-sponse, shown in Figure 14. Figures of merit calculated for a l l possible sources are tested against these results. The signal is rejected i f the figure of merit is more than 10% below that expected for a true unresolved source with the equivalent rms signal energy. Flux densities and epoch 1950 co-ordinates are computed, for these sources which pass this test, based on the position in the scan, peak to peak signal amplitude and positive to negative peak ratio. The program is executed using data averaged over a l l days and both re-ceivers. This extensive averaging reduces the rms receiver noise by a,-factor /40 to about 1 mJy. Thus, the detection sensitivity is limited only by the rms confusion level in each scan. The program yields a catalogue of compact sources at 5 GHz in the region of the galactic plane covered by the observations. 42. The portion of the program to search for var i a b i l i t y in the observa-tions i s not yet complete. Two approaches to this study have been consid-ered; depending on the scale of intensity variation to be measured. The measurement of fractional variations in strong sources, which are present on a l l days of the survey well above the rms confusion level, is best carried out by the simple method of running the detection program, just outlined, on each day independently and measuring deviations from the average intensity. This method takes advantage of the high signal to noise of these sources which allow the effects of daily telescope pointing fluctua-tions to be removed (see section B.). In the case of v a r i a b i l i t y of weaker sources (less than ^30 mJy), for which the unresolved source profile may be heavily distorted by confusion, and for highly variable sources which may appear on only a few days of the survey, a more suitable approach is one similar to that used in the 21 cm survey of differencing each day from the average. This method subtracts out, to a large degree, the effects of these confusion sources which are stable over the complete observation period. It should be noted that for these weak sources the survey sensitivity enables only variations which correspond to a large percentage change in the i n t r i n -sic source emission to be detected. 90 Oi LU S °70f to oi 60 50h" I 1 J 1 1 —I . I i L 5 10 15 20 25 30 35 40 SIGNAL TO NOISE Figure 14. Figure of Merit versus Rms Signal to Noise 4 4 . D. Survey Results and GT 0 2 3 6 + 6 1 0 The results of the survey for compact sources in the observations are presented in Table IV. A total of 1 9 6 sources were detected. Column 1 l i s t s the survey source names. Based on the recommendation of Commission 28 of the International Astronomical Union (see Kesteven and Bridle, 1 9 7 7 ) the "eight d i g i t " source name has been adopted. The f i r s t 4 digits indicate hours and minutes of right ascension and the last 4 the sign and degrees of declination, truncated to a tenth of a degree, for the epoch 1 9 5 0 . Columns 2 and 3 give the computed right ascension and declination of the source. The source intensity in mJy is listed in Column 4 . Conversion from antenna temperature to absolute flux introduces an uncertainty of ^10% in the abso-lute source strength. Some of the stronger sources may be identified with sources detected in previous surveys. These identifications are listed in Column 5 of the table. In the case of identification with sources for which precise positions are available, the accuracy of the calculated positions can be determined. On the basis of 1 0 such comparisons, spread over the declination range 2 5 ° - H J 3 ° , a two sigma error of 6 seconds in right ascen-sion and 1 5 " in declination may be placed on the source co-ordinates quoted in the table. Although the vari a b i l i t y study, as mentioned, is s t i l l under develop-ment, one highly variable source was discovered by direct inspection of the strip chart recordings of the analogue receiver outputs. The results of a more detailed analysis of the data for this radio source, designated GT 0 2 3 6 + 6 1 0 , and i t s possible association with the COS B y-vay source CG 1 3 5 + 1 (Hermsen et a l . , 1 9 7 7 ) have been published elsewhere (Gregory and Taylor, 1 9 7 8 ) . To summarize briefly, during this survey the source experi-45. Table IV: A l i s t of Compact Sources in the 6 cm Survey .Region (1) Name (2); a(1950) (3) (4) (5) 6(1950) Flux (mJy) Identification GT 1915+129 19 h ,_m 15 23 S 12° 54* 09" 22 1927+156 19 27 50 15 40 42 16 1933+188 19 33 06 18 51 57 43 1933+181 19 33 34 18 11 22 17 1934+200 19 34 47 20 04 49 14 1935+227 19 35 52 22 42 04 44 1938+215 19 38 31 21 33': 15 35 1939+207 19 39 00 20 42 19 14 1940+222 19 40 22 22 14 37 15 1941+252 19 41 36 25 14 55 19 1945+241 19 45 51 24 07 33 143 1945+247 19 45 56 24 42 03 99 1948+281 19 48 30 28 09 43 24 1950+250 19 50 05 25 04 04 21 1952+287 19 52 27 28 47 59 38 1959+321 19 59 16 32 08 35 71 1959+312 19 59 47 31 13 18 50 2000+297 20 00 34 29 45 49 62 2000+287 20 00 50 28 45 38 13 2002+313 20 02 42 31 18 43 39 2003+326 20 03 20 32 41 15 16 2008+343 20 08 38 34 22 40 27 2010+349 20 10 13 34 59 04 22 2012+330 20 12 31 33 04 28 25 2018+355 20 18 12 35 35 26 21 2018+359 20 18 23 35 58 35 12 2023+372 20 23 28 37 13 22 1320 2041+434 20 41 26 43 28 01 40 2059+465 20 59 25 46 33 24 24 2101+472 21 01 34 47 14 01 16 2102+463 21 02 01 46 22 16 11 2109+480 21 09 . 24 48 04 45 14 2111+494 21 11 27 49 24 57 26 2112+479 21 12 11 47 56 07 25 2114+495 21 14 32 49 35 34 140 2115+511 21 15 47 15 08 54 78 2116+493 21 16 35 49 23 15 22 2119+493 21 19 03 49 22 29 11 2119+504 21 19 26 50 25 51 20 2126+503 21 26 40 50 18 52 26 B2.2 1945+24 B2 2000+29 B2.3 2023+37 BG 2114+49 LHE 500 Table IV (continued) (1) (2) (3) (4) GT 2127+487 21 h 27 m 31 S 48° 46' 17" 24 2132+498 21 32 59 49 52 25 23 2134+531 21 34 21 53 11 11 80 2134+536 21 34 28 53 38 27 10 2136+526 21 36 14 52 36 44 33 2138+526 21 38 19 52 36 44 74 2138+533 21 38 58 53 24 50 22 2140+530 21 40 52 53 00 22 17 2141+521 21 41 24 52 07 58 22 2142+517 21 42 45 51 45 59 12 2143+532 21 43 34 53 15 38 17 2145+542 21 45 45 54 16 25 23 2147+522 21 47 21 52 12 29 18 2148+550 21 48 40 55 02 04 43 2148+553 21 48 56 55 19 57 18 2152+530 21 52 20 53 03 27 27 2155+553 21 55 08 55 20 33 20 2156+531 21 56 07 53 09 25 43 2157+554 21 57 56 55 26 51 59 2202+547 22 02 12 54 46 00 11 2206+454 22 06 33 45 29 41 18 2211+554 22 11 44 55 25 08 12 2211+558 22 11 51 55 50 51 24 2212+560 22 12 08 56 05 32 33 2214+564 22 14 39 56 26 29 33 2217+579 22 17 13 57 55 34 18 2220+559 22 20 55 55 55 20 25 2223+573 22 23 53 57 19 55 42 2228+587 22 28 01 58 42 13 100 2234+568 22 34 07 56 51 46 51 2235+585 22 35 34 58 30 44 75 2236+596 22 36 06 59 36 05 29 2236+600 22 36 26 60 04 29 32 2237+599 22 37 22 59 56 24 77 2238+572 22 38 27 57 17 25 160 2240+580 22 40 00 58 04 06 45 2240+583 22 40 12 58 28 06 14 2242+568 22 44 16 56 52 31 41 2248+572 22 48 42 57 12 46 23 2249+592 22 49 29 v 59 14 26 51 2249+597 22 49 53 59 43 54 15 2251+604 22 51 45 60 27 42 310 Table IV (continued) CD (2) (3) (4) (5) GT 2251+598 22 h 5 l m 50 S 59 2251+595 22 51 55 59 2252+591 22 52 21 59 2252+576 22 52 57 57 2255+606 22 55 03 60 2257+577 22 57 51 57 2259+610 22 59 38 61 2259+616 22 59 56 61 2300+616 23 00 25 61 2301+597 23 01 20 59 2303+603 23 03 48 60 2305+612 23 05 33 61 2305+604 23 05 53 60 2306+591 23 06 30 59 2308+612 23 08 46 61 2310+615 23 10 11 61 2324+618 23 24 11 61 2324+611 23 . 24 23 61 2329+605 23 29 37 60 2329+598 22 29 43 59 2330+595 23 30 47 59 2335+603 23 35 26 60 2341+630 23 41 09 63 2342+618 23 42 53 61 2343+609 23 43 21 60 2343+604 23 43 30 60 2344+609 23 44 51 60 2345+615 23 45 15 61 2345+633 23 45 46 63 2345+635 23 45 47 63 2347+609 23 47 54 60 2349+606 23 49 11 60 2552+626 .23 52 19 63 2352+602 23 52 55 60 2353+613 22 53 51 61 2354+611 23 54 06 61 ' 2357+613 23 57 09 61 0001+628 00 01 06 62 0001+613 00 01 49 61 0007+602 00 07 18 60 0007+614 00 07 41 61 49' 09" 32 32 56 22 06 22 36 40 59 1370 BG 2252+57 40 46 25 42 47 89 01 00 27 38 31 15 39 27 35 42 14 17 19 41 19 15 59 23 28 30 16 09 02 36 15 12 96 35 03 15 32 23 46 06 19 20 30 01 13 51 17 17 34 34 28 21 22 100 00 19 33 50 21 24 59 51 30 26 52 16 59 14 90 OZ+674 35 00 44 19 37 19 32 45 26 55 36 34 40 02 170 OZ+682 31 51 26 14 59 60 22 33 31 08 27 16 21 45 17 51 35 38 21 27 37 14 06 29 25 30 10 Table IV (continued) (1) (2) (3) (4) GT 0008+626 oo h 08 m 30 S 62° 40* 37" 19 0010+627 00 10 31 62 46 22 69 0011+610 00 11 04 61 01 13 16 0011+605 00 11 56 60 35 15 18 0012+610 00 12 02 61 01 02 2100 0013+637 00 13 16 63 44 47 18 0015+614 00 15 40 61 28 28 23 0017+627 00 17 18 62 45 07 24 0017+630 00 17 40 63 03 22 19 0018+641 00 18 58 64 10 32 30 0018+639 00 18 52 63 54 48 19 0020+606 00 20 31 60 37 27 15 0025+605 00 25 04 60 31 01 51 0025+613 00 25 53 61 18 15 18 0026+627 00 26 23 62 46 58 45 0026+637 00 26 47 63 42 12 1060 0028+631 00 28 43 63 06 23 84 0028+625 00 28 52 62 33 26 32 0030+610 00 30 07 61 02 26 22 0030+618 00 30 30 61 53 49 17 0030+622 00 30 51 62 14 36 16 0040+634 00 40 31 63 25 15 16 0040+637 00 40 38 63 43 48 28 0053+624 00 53 30 62 25 41 46 0056+628 00 56 08 62 50 57 27 0056+606 00 56 53 60 37 26 31 0057+613 00 57 29 61 22 02 24 0100+628 01 00 27 62 52 22 18 0101+615 01 01 09 61 33 14 31 0102+622 01 02 32 62 13 30 39 0103+628 01 03 06 62 52 53 38 0103+631 01 03 09 63 10 38 140 0104+630 01 04 53 63 00 57 14 0105+614 01 05 25 61 28 21 22 0109+631 01 09 26 63 10 03 15 0111+615 01 11 12 61 33 05 17 0114+610 01 14 50 61 00 46 16 0116+635 01 16 50 63 30 20 22 0119+615 01 19 14 61 35 15 82 0120+610 01 20 30 61 01 05 38 0122+639 01 22 42 63 57 53 56 0123+616 01 23 26 61 37 03 32 0127+637 01 27 09 63 46 46 35 (5) DW 0012+61 3C 11.1 BG 0103+63 49. Table IV (continued) (1) (2) (3) (4) GT 0141+625 oi h 41 m ios 62° 35' 18" 35 0144+637 01 44 38 63 45 25 24 0147+600 01 47 38 60 00 02 12 0148+610 01 48 14 61 02 21 48 0158+623 01 58 12 62 23 56 54 0159+634 01 59 21 63 25 41 64 0202+611 02 02 19 61 10 17 20 0207+616 02 07 09 61 38 20 14 0207+628 02 07 27 62 49 17 16 0208+626 02 08 16 62 36 03 32 0208+621 02 08 29 62 10 20 31 0209+596 02 09 54 59 41 49 13 0210+605 02 10 50 60 34 21 16 0212+629 02 12 16 62 57 13 33 0215+601 02 15 44 60 11 07 17 0216+626 02 16 56 62 37 09 23 0224+584 02 24 07 58 27 05 35 0229+591 02 29 15 59 10 59 30 0231+620 02 31 32 62 03 17 75 0232+594 02 32 46 59 26 12 15 0234+589 02 34 14 58 59 04 790 0236+604 02 36 52 60 24 40 18 0238+584 02 38 46 58 27 30 50 0242+580 02 42 20 58 00 22 28 0243+583 02 43 12 58 21 57 18 0251+585 02 51 05 58 21 47 29 0256+565 02 56 57 56 33 25 22 0257+582 02 57 44 58 16 20 27 0301+564 03 01 46 56 27 15 32 0307+588 03 07 46 58 46 29 24 (5) WE 0232+59W1 3C 069 50. enced a radio outburst which was f i r s t detected on Aug. 24 and reached a maximum flux density of 285 mJy on Aug. 27. The intensity slowly decreased during the remaining 3 days of the survey, so that by Aug. 30 i t was at a level of 170 mJy. In collaboration with Drs. Hjellming, Hogg and Hvatum, we obtained a precise radio position of the source, using the Very Large Array in New Mexico, at 4885 MHz (Hjellming et a l . , 1978). a(1950) = 02 h 36 m 40?62 ± 0?14 6(1950) = 61° 00' 54" ± 1" These precise co-ordinates confirmed the tentative identification of GT 0236+610 with the star LSI61°303 (Gregory and Taylor, 1978; Sanduleak, 1978). Crampton and Hutchings (1978) classify this star as Bllb and note the presence of broad H^ and H^  emission lines, with total width ^900 km/sec, superimposed on sharper central absorbtion. UBV photometry (Rossiger, 1978) gives V = 10.72, B = 11.54 and U = 11.29. The large fractional variations observed with the one day sample time of this survey implies an upper limit to the source radius of ^2.5 X 10"^ cm. This limit was further refined by higher time resolution observations, at 2.8 cm, of GT 0236+610; undertaken by Dr. Gregory and myself during Feb. 18->March 1, 1978 with the 46 meter telescope of the Algonquin Radio Observatory. During the interval Feb. 18->23 the source was not present, and data taken during this period were used to determine the confusion at the source position as a function of hour angle. Figure 15. shows the observed source intensity for the period Feb. 24 to March 31. The source appeared to suddenly turn on at about 12 hours U.T. on Feb. 24. Thereafter, the intensity increased from day to day in an almost "step-wise" manner; remaining essentially constant for long periods of time and then suddenly increasing by up to 100%, on a time scale of ^ 2 hrs. This time scale for variation reduces the upper limit on the radius of the emission region to ^ 1011+ cm; similar in size to Cyg X-3 (see Introduction). In general, the results of this suvery confirm the conclusion of the 21 cm survey that short term intensity variations at a level of 1 Jansky or greater are exceedingly rare. The well established detection of one highly variable source strengthens the possibility that weaker variations are more common. A complete answer to this question must await completion of the var i a b i l i t y survey. As a final note concerning the compact source survey; i f the detection of 196 sources in the 56 sq. deg. coverage of this survey is taken to be typical of the galactic plane, a mean source density of ^ IO4 ster 1 is implied. Extrapolation of source counts at 5 GHz (Kellermann et a l . , 1971) to 10 mJy, assuming a Euclidean power law of -1.5, yields the expected value of 5 x I0k ster The apparent deficiency of detections in thissurvey can be attributed to the fact that the sensitivity is a function of confusion, and hence varies from place to place in the survey region. Thus, although the catalogue of compact sources given in Table IV has a minimum detection limit of 10 mJy, i t is not complete down to this level. Figure 15. Observations of GT 0236+610 at 2.8 cm 52. 100 50 150 100 6 >-z PJ Q X •J 50 dill I L .12 Feb. 24 18 • • • • • • 12 hrs. UT Feb. 25 _i_ 150 100 50 h 12 Feb. 26 18 12 UT Feb. 27 J L 12 18 Feb. 28 12 March 1 UT D. Summary and Conclusions In the Introduction i t was stated that the purpose of this invest-gation was to attempt to answer the question: How time variable is the radio sky? We are now in a position to provide at least some partial answers to this question. The stipulation that highly variable sources be both physically compact and detectable essentially restricts a search for these sources to our own galaxy. Thus, both of these surveys are centered on the galactic plane, where the highest concentration would be expected.to be located. No variations with magnitude greater than 1 Jansky were observed during the 21 cm survey. Exam-ination of the strip chart recordings also seems to rule out variations on this scale in the 6 cm survey. These facts demonstrate clearly that variations of this magnitude are very rare. The situation with regard to weaker variations is less clear; but the tentative detection of one variation of ^ 0.9 Jy at 21 cm and the confirmed detection of at least one highly variable source at a level of 0.3 Jy at 6 cm, strongly suggests that the prospect of future detection of variability below 1 Jansky is much more promising. More concrete data on smaller scale variations w i l l be available on completion of the 6 cm var i a b i l i t y study. A search Of the average of the 6 cm survey data has revealed 196 com-pact radio sources in the 56 sq. deg. area coverage. This total implies a mean density of compact source detections in this area of the plane of about 10^ ster 1 , above 10 - 20 mJy. The planned complete survey of the galactic plane above 6=0° may be expected to yield a catalogue of ^4000 sources. Previous sky surveys have been restricted to single beam radiometers i n order to accurately chart the complex radio structure i n the g a l a c t i c plane. As a r e s u l t , these surveys have been li m i t e d to minimum source i n t e n s i t i e s of ^ 200 mJy by the large dynamic range of the confusing structure. The r e s u l t s of the 6 cm survey c l e a r l y demonstrate- the feas b i l i t y of a beamswitching technique i n a survey scale endeavour, and i t s value i n detecting low l e v e l compact sources amidst large confusing structure. 55. Bibliography 1. Bridle, A. H., Kesteven, M. J. K., and Brandie, G. W., Astron. J., 82_, 21 (1977) . 2. Crampton, D., and Hutchings, J. B., IAU Circ. No. 3180 (1978). 3. Dent, W. A., Science, 148, 1458 (1965). 4. Feller, W., An Introduction to Probability Theory and Its Application, Vol. I (Wiley Publications in Statistics, 1950). 5. Geoffrey, T. B., and Wallerstein, G., PASP, 88_, 759 (1976). 6. Gregory, P. C., Kronberg, P. P., Seaquist, E. R., Hughes, V. A., Woodsworth, A., Viner, M. R., and Retallack, D., Nature, 239, 440 (1972a). 7. Gregory, P. C., Kronberg, P. P., Seaquist, E. R., Hughes, V. A., Woodsworth, A., Viner, M. R., Retallack, D., Hjellming, R. M., and Balick, B., Nature Physical Science, 239, 114 (1972b). 8. Gregory, P. C., and Seaquist, E. R., Ap. J., 194, 715 (1974). 9. Gregory, P. C., and Taylor, A. R., Nature, 272, 704 (1978). 10. Hermsen, W., Swanenburg, B. N., Bignami, G. F., Boella, G., Buccheri, R., Scarsi, L., Kanbach, G., Mayer-Hasselwander, H. A., Masnou, J. L., Paul, J. A., Bennett, K., Higdon, J. C., Lichti, G. G., Taylor, B. G., and Wills, R. D., Nature, 269, 494 (1977). 11. Hewish, A., Bell, S. J., Pilkington, J. D. H., Scott, P. F., and Collins, R. A., Nature, 217, 709 (1968). 12. -Hjellming, R., Hogg, D., Hvatum, H., Gregory, P., and Taylor, R., IAU Circ. No. 3180 (1978). 13. Jaschek, C , and Gomez, A., PASP, 82, 809 (1970). 14. Kellermann, K. I., and Pauliny-Toth, I. I. K., Ann. Rev. of Astron. and Astrophys., 6_, 417 (1968). 15. Kellermann, K. I., in Galactic and Extra-Galactic Radio Astronomy, ed. Gerrit L. Vershuur and Kenneth I. Kellermann (Springer-Verlag New York, 1974) p. 320. 16. Kellermann, K. I., Davis, M. M., and Pauliny-Toth, I. I. K., Ap. J., 170, LI (1971). 56. 17. Kesteven, M. J. L., Bridle, A. H., and Brandie, G. W., Astron. J., 81_, 919 (1976). 18. Kesteven, M. J. L., Bridle, A. H., and Brandie, G. W., Astron. J., 82, 541 (1977). 19. Kesteven, M. J. L., and Bridle, A. H., JRASC, 71_, 21 (1977). 20. Owen, F. N., Jones, T. W., and Gibson, D. M., Ap. J., 210, L27 (1976). 21. Rossiger, S., IAU Circ. No. 3210 (1978). 22. Sanduleak, N., IAU Circ. No. 3170 (1978). 23. Seaquist, E. R., Gregory, P. C , Perley, R. A., Becker, R. H., Carlson, J. B., Kundu, M. R. , Bignell, R. C , and Dickel, J. R. , Nature, 251, 394 (1974) . 24. Spangler, S. R., Owen, F. N., and Hulse, R. A., Astron. J., 8_2, 989 (1977). 25. Tu i r i , E. M., in Radio Astronomy, by John D. Krauss (McGraw-Hill, 1966) p. 258. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0094492/manifest

Comment

Related Items